Empirical Evidence on Audit Quality under a Dual Mandatory Auditor Rotation Rule

Abstract

Regulators in the US ruled against introducing mandatory firm rotations in addition to the existing rule for periodic partner rotations. In contrast, European regulators ruled in favor of a dual mandatory rotation rule in which both audit firm and audit partner rotations are required. Employing a unique setting where a dual regime of audit and firm rotations are required, we assess the net benefit (cost), of audit firm rotation incrementally to partner rotation. Specifically, we analyze several earnings-based measures of audit quality along with the market perception of audit quality. Controlling for partner rotation, we do not find that firm rotations have a positive incremental effect. In contrast, we find audit partner rotation under the dual regime appears to improve both the earnings-based measures of audit quality, and market perceptions of earnings. Our evidence suggests that any benefit arising from dual rotation is likely to be driven by the change in partner. However, whether the audit firm rotation should still be required is unclear, given that the observed benefits arising from the audit partner rotation could potentially be preconditioned on audit firm rotation.

Keyword:

• Audit-firm rotations
• Audit-partner rotations
• Audit quality
• Investor perceptions

Acknowledgements

We thank Henrik Nilsson (editor) and the two anonymous reviewers, Julie Barrow, Jere Francis, Fani Kalogirou, Clive Lennox, Kevin McMeeking, and workshop and seminar participants at Exeter Business School, American Accounting Association, and Nanyang Technological University in Singapore.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Supplemental Data and Research Materials

Supplemental data for this article can be accessed at https://doi.org/10.1080/09638180.2020.1747513.

Table A1. Table 3 – Mandatory rotation and reporting quality with extension to earlier and later years (2006–2012)

Table A2. Table 5 – Mandatory audit firm rotation and reporting quality during single rotation regime (1993–2005)

Table A3. Table 6 – Mandatory audit firm rotations and reporting quality pre- and post-dual audit rotation regime (1993–2012)

Table A4. Table 3 – Mandatory rotation and reporting quality: sensitivity analysis to the exclusion of all client firms experiencing a voluntary partner or firm rotation during 2006–2012, or alternatively, controlling for voluntary rotations

Table A5. Table 5 – Mandatory rotation and reporting quality: sensitivity analysis to the exclusion of all client firms experiencing a voluntary partner or firm rotation during 1993–2005, or alternatively, controlling for voluntary rotations

Table A6. Table 6 – Mandatory rotation and reporting quality: sensitivity analysis to the exclusion of all client firms experiencing a voluntary partner or firm rotation during 1993-2012, or alternatively, controlling for voluntary rotations

Table A7. Firm FE but without clustering

Table A8. Clustering by firm with industry fixed effects

Table A9. Signed AQ measures

Table A11. Chen et al. (2018)

Table A12. Aobdia (2019)

ORCID

Gilad Livne http://orcid.org/0000-0003-2270-6453

Angela Pettinicchio http://orcid.org/0000-0001-8845-5066

Notes

1 REGULATION (EU) No 537/2014. This regulation applies to public interest entities (PIEs), which essentially are publicly listed firms. Thus, the new regulation can be viewed as an attempt by the EU to increase oversight and monitoring of firms whose shares are widely held by the public.

2 Independence in-fact is defined a state of mind that is unaffected by influences that might compromise professional judgement and allows an individual to act with integrity and or exercise objectivity and professional scepticism (IFAC, 2018).

3 The literature on audit tenure indicates higher rate of audit failures during the initial engagement period (Geiger & Raghunandan, 2002) and that audit quality is lower in the first two-three years (Johnson et al., 2002).

4 The limit was then extended to seven years in 2013.

5 This is consistent with Lennox et al. (2014, p. 1777) who refer to the embarrassment effect as follows: ‘[T]he departing partner has an incentive to conduct a higher quality audit in his/her final year t in order to avoid the embarrassment of the audit deficiencies being found by the incoming partner in year t+1.’ The embarrassment effect thus leads to higher audit effort by the outgoing auditor, which further mitigates audit risk.

6 The international Federation of Accountants (IFAC) defines independence as follows: ‘Independence of mind – the state of mind that permits the expression of a conclusion without being affected by influences that compromise professional judgement, thereby allowing an individual to act with integrity, and exercise objectivity and professional skepticism. Independence in appearance – the avoidance of facts and circumstances that are so significant that a reasonable and informed third party would be likely to conclude that a firm’s or an audit or assurance team member’s integrity, objectivity or professional skepticism has been compromised.’ (See, https://www.ifac.org/system/files/publications/files/Final-Pronouncement-The-Restructured-Code_0.pdf.) Dopuch et al. (2003, p. 84) provides definitions similar in spirit although he uses the term ‘independence in-fact’ and not ‘independence of mind.’

7 Consistent with this, Cameran et al. (2015) provide evidence that audit fee and effort are higher prior to mandatory firm rotations. However, as we report later, we do not find evidence consistent with improved reporting quality in the last year of the audit engagement.

8 See http://pcaobus.org/rules/rulemaking/pages/docket037comments.aspx, comment letter No. 136.

9 See http://ec.europa.eu/internal_market/auditing/reform/.

10 For a more detailed list of countries in which the mandatory rotation rule is enforced, see Lennox (2014).

11 Both studies in robustness tests find tenure is associated with audit quality; longer tenure is associated with higher levels of conservatism.

12 Voluntary auditor changes may be caused by a variety of reasons including the health of the client-firm, the need to realign the needs of managers, severity of audit opinion and overly conservative auditors (See DeFond & Zhang, 2014, for a review of the literature).

13 In particular, a break was mandated in 2006 for all partner engagements which in that year had a duration of six years or more.

14 See Cameran (2005) for further detail on the Italian audit market.

15 Audit firm data are also available on Compustat, but many mistakes were found. For this reason, audit firm identity was manually checked against the client’s annual report.

16 Using the Fama-French 12-industries classification.

17 We find our results are not sensitive to the exclusion of voluntary rotation firm-years. See section 5 below.

18 Using current loss does not change our results.

19 The mean and median values of our leverage ratio are lower than what is reported by Cameran et al. (2016) (0.53 & 0.55, respectively). However, we account only for debt liabilities whereas Cameran et al. (2016) use total liabilities. Growth rate in their sample is slightly higher than ours (mean: 0.11; median: 0.07).

20 We run the VIF test in our regressions and results show no sign of multicollinearity issues (maximum VIF around 2, well below the 10-threshold suggested by Kennedy, 2008).

21 Most of the controls in Cameran et al. (2015) in their analysis of AAWCA are also insignificant, as they also use firm fixed effects.

22 The following variables lose significance in one or more specifications once we include firm fixed effects instead of industry fixed effects: SIZE, AGE, ROA, CFO, AGE*ROA and the intercept

23 This analysis focuses on the immediate effects of firm and partner rotations relative to the previous year, and so cannot speak about longer-term effects of rotations. Note that, like Cameran et al. (2016) (in the case of mandatory firm rotations), this specification holds fixed the client firm. However, unlike Cameran et al. (2016), it compares the year audited by the outgoing audit firm to the year audited by the incoming audit firm.

24 We excluded year 2005 as this was the first year in which IFRS were introduced for listed companies in Italy.

25 Our results remain the same if we control for partner voluntary rotations in this period.

26 The online appendix also reports other analyses that, for brevity, are not discussed in this section.